基于间接健康特征优化与多模型融合的锂电池SOH-RUL联合预测

doi:10.19595/j.cnki.1000-6753.tces.231057

Abstract
Figure/Table
References
Related Citation (12)

Download: PDF (27269 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract Lithium-ion batteries have become essential for new energy vehicles and energy storage power stations due to their high energy density, low self-discharge rate, and reliable cleanliness. However, lithium batteries gradually age with increasing cycle counts, leading to decreased battery performance. State of health (SOH) and remaining useful life (RUL) are important indicators for evaluating battery aging status, and accurate predictions of these metrics are crucial for the safe operation of energy storage systems. However, choosing suitable indirect health features (IHF) for SOH and RUL predictions is challenging. Data-driven models generate uncertainties, resulting in inaccurate predictions of SOH and RUL. Therefore, this paper proposes a joint prediction method for lithium battery SOH and RUL based on indirect health feature optimization and multi-model fusion.
Firstly, multiple health factors (HF) are extracted from charging voltage curves, and IHF are obtained through feature concatenation and attention mechanism optimization, denoted as feature Co-HF. Then, Bayesian model averaging (BMA) is introduced to address uncertainties in the prediction process. With support vector regression (SVR) and long short-term memory (LSTM), SVR-BMA and LSTM-BMA models are constructed. Additionally, the input features for RUL prediction are extracted from SOH capacity predictions using the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) method, enabling joint prediction of SOH and RUL.
The proposed method is validated using the University of Maryland's CALCE lithium battery dataset. Compared with different HFs and predictions of each sub-model, the Co-HF capacity prediction results are closer to the measured values, and the tracking effect of capacity trends is ideal. The mean absolute error (MAE) and root mean squared error (RMSE) of SOH predictions using the proposed fusion model are 0.008 1 and 0.012 0, respectively, and the MAE and RMSE of RUL predictions are 2.56 and 4.63, respectively. The prediction errors are lower than those of each sub-model. In addition, the error tolerance of SOH and RUL predictions is within 5% at SP=100/200/300. Thus, the proposed method significantly reduces the MAE and RMSE of SOH and RUL predictions.
From the experimental analysis, it can be concluded that: (1) the optimized feature Co-HF comprehensively reflects the aging information in lithium battery charging voltage, which is suitable for SOH predictions. (2) The proposed model effectively reduces prediction errors of lithium battery SOH and RUL, improving the accuracy and reliability of predictions. Future work will focus on selecting indirect health features that reflect special battery capacity degradation laws and adopting better algorithms to solve model parameters.

Key words： State of health remaining useful life indirect health factor bayesian model averaging support vector regression long short-term memory

Received: 05 July 2023

PACS:

TM912

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	Cai Yusi
	Li Zewen
	Liu Ping
	Xia Xiangyang
	Wang Wen

Cite this article:

Cai Yusi,Li Zewen,Liu Ping等. Joint Prediction of Lithium Battery State of Health and Remaining Useful Life Based on Indirect Health Features Optimization and Multi-Model Fusion[J]. Transactions of China Electrotechnical Society, 2024, 39(18): 5883-5898.

URL:

https://dgjsxb.ces-transaction.com/EN/10.19595/j.cnki.1000-6753.tces.231057 OR https://dgjsxb.ces-transaction.com/EN/Y2024/V39/I18/5883

[1] 李先锋, 张洪章, 郑琼, 等. 能源革命中的电化学储能技术[J]. 中国科学院院刊, 2019, 34(4): 443-449.
Li Xianfeng, Zhang Hongzhang, Zheng Qiong, et al.Electrochemical energy storage technology in energy revolution[J]. Bulletin of Chinese Academy of Sciences, 2019, 34(4): 443-449.
[2] 牛志远, 姜欣, 谢镔, 等. 电动汽车过充燃爆事故模拟及安全防护研究[J]. 电工技术学报, 2022, 37(1): 36-47, 57.
Niu Zhiyuan, Jiang Xin, Xie Bin, et al.Study on simulation and safety protection of electric vehicle overcharge and explosion accident[J]. Transactions of China Electrotechnical Society, 2022, 37(1): 36-47, 57.
[3] Hu Xiaosong, Deng Xinchen, Wang Feng, et al.A review of second-life lithium-ion batteries for stationary energy storage applications[J]. Proceedings of the IEEE, 2022, 110(6): 735-753.
[4] Deng Yuanwang, Ying Hejie, Jiaqiang E, et al.Feature parameter extraction and intelligent esti- mation of the state-of-health of lithium-ion batte- ries[J]. Energy, 2019, 176: 91-102.
[5] Meng Jinhao, Cai Lei, Luo Guangzhao, et al. Lithium-ion battery state of health estimation with short-term current pulse test and support vector machine[J]. Microelectronics Reliability, 2018, 88/ 89/90: 1216-1220.
[6] Yang Duo, Zhang Xu, Pan Rui, et al.A novel Gaussian process regression model for state-of-health estimation of lithium-ion battery using charging curve[J]. Journal of Power Sources, 2018, 384: 387-395.
[7] Yayan U, Arslan A T, Yucel H.A novel method for SOH prediction of batteries based on stacked LSTM with quick charge data[J]. Applied Artificial Intel- ligence, 2021, 35(6): 421-439.
[8] 肖迁, 穆云飞, 焦志鹏, 等. 基于改进LightGBM的电动汽车电池剩余使用寿命在线预测[J]. 电工技术学报, 2022, 37(17): 4517-4527.
Xiao Qian, Mu Yunfei, Jiao Zhipeng, et al.Improved LightGBM based remaining useful life prediction of lithium-ion battery under driving conditions[J]. Transactions of China Electrotechnical Society, 2022, 37(17): 4517-4527.
[9] 赵靖英, 胡劲, 张雪辉, 等. 基于锂电池模型和分数阶理论的SOC-SOH联合估计[J]. 电工技术学报, 2023, 38(17): 4551-4563.
Zhao Jingying, Hu Jin, Zhang Xuehui, et al.Joint estimation of the SOC-SOH based on lithium battery model and fractional order theory[J]. Transactions of China Electrotechnical Society, 2023, 38(17): 4551-4563.
[10] Park J, Lee M, Kim G, et al.Integrated approach based on dual extended Kalman filter and multivariate autoregressive model for predicting battery capacity using health indicator and SOC/SOH[J]. Energies, 2020, 13(9): 2138.
[11] Zhang Yongzhi, Xiong Rui, He Hongwen, et al.Long short-term memory recurrent neural network for remaining useful life prediction of lithium-ion batteries[J]. IEEE Transactions on Vehicular Tech- nology, 2018, 67(7): 5695-5705.
[12] 王萍, 弓清瑞, 张吉昂, 等. 一种基于数据驱动与经验模型组合的锂电池在线健康状态预测方法[J]. 电工技术学报, 2021, 36(24): 5201-5212.
Wang Ping, Gong Qingrui, Zhang Jiang, et al.An online state of health prediction method for lithium batteries based on combination of data-driven and empirical model[J]. Transactions of China Electro- technical Society, 2021, 36(24): 5201-5212.
[13] Liu Kailong, Shang Yunlong, Ouyang Quan, et al.A data-driven approach with uncertainty quantification for predicting future capacities and remaining useful life of lithium-ion battery[J]. IEEE Transactions on Industrial Electronics, 2020, 68(4): 3170-3180.
[14] Vosseler A, Weber E.Forecasting seasonal time series data: a Bayesian model averaging approach[J]. Computational Statistics, 2018, 33(4): 1733-1765.
[15] Yang Jing, Fang Gonghuan, Chen Yaning, et al.Climate change in the Tianshan and northern Kunlun mountains based on GCM simulation ensemble with Bayesian model averaging[J]. Journal of Arid Land, 2017, 9(4): 622-634.
[16] 王萍, 范凌峰, 程泽. 基于健康特征参数的锂离子电池SOH和RUL联合估计方法[J]. 中国电机工程学报, 2022, 42(4): 1523-1534.
Wang Ping, Fan Lingfeng, Cheng Ze.A joint state of health and remaining useful life estimation approach for lithium-ion batteries based on health factor parameter[J]. Proceedings of the CSEE, 2022, 42(4): 1523-1534.
[17] Yu Jianbo.State of health prediction of lithium-ion batteries: multiscale logic regression and Gaussian process regression ensemble[J]. Reliability Engin- eering & System Safety, 2018, 174: 82-95.
[18] 黄凯, 丁恒, 郭永芳, 等. 基于数据预处理和长短期记忆神经网络的锂离子电池寿命预测[J]. 电工技术学报, 2022, 37(15): 3753-3766.
Huang Kai, Ding Heng, Guo Yongfang, et al.Prediction of remaining useful life of lithium-ion battery based on adaptive data preprocessing and long short-term memory network[J]. Transactions of China Electrotechnical Society, 2022, 37(15): 3753-3766.
[19] Boulahia S Y, Amamra A, Madi M R, et al.Early, intermediate and late fusion strategies for robust deep learning-based multimodal action recognition[J]. Machine Vision and Applications, 2021, 32(6): 121.
[20] 戴俊彦, 夏明超, 陈奇芳. 基于双重注意力机制的电池SOH估计和RUL预测编解码模型[J]. 电力系统自动化, 2023, 47(6): 168-177.
Dai Junyan, Xia Mingchao, Chen Qifang.Encoding and decoding model of state of health estimation and remaining useful life prediction for batteries based on dual-stage attention mechanism[J]. Automation of Electric Power Systems, 2023, 47(6): 168-177.
[21] Zhi Yue, Wang Heqi, Wang Liang.A state of health estimation method for electric vehicle Li-ion batteries using GA-PSO-SVR[J]. Complex & Intelligent Systems, 2022, 8(3): 2167-2182.
[22] Sayah M, Guebli D, Al Masry Z, et al.Robustness testing framework for RUL prediction deep LSTM networks[J]. ISA Transactions, 2021, 113: 28-38.
[23] Lü Zhiqiang, Wang Geng, Gao Renjing.Li-ion battery prognostic and health management through an indirect hybrid model[J]. Journal of Energy Storage, 2021, 42: 102990.
[24] Bak T, Lee S.Accurate estimation of battery SOH and RUL based on a progressive LSTM with a time compensated entropy index[C]//Annual Conference of the PHM Society, Scottsdale, 2019: 2012-2013.